
import pandas as pd

from sklearn.preprocessing import StandardScaler

import seaborn as sns 
import matplotlib.pyplot as plt

from scipy.spatial.distance import cdist
from sklearn.cluster import KMeans
import numpy as np 

from scipy.spatial.distance import pdist
from scipy.cluster.hierarchy import dendrogram, linkage,cophenet


freedom_df = pd.read_csv('C:\\Anil\\AI And ML\\economic_freedom_index2019_data.csv', encoding='ISO-8859-1')
freedom_df.shape

(186, 34)


freedom_df.head()


freedom_df.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 186 entries, 0 to 185
Data columns (total 34 columns):
 #   Column                       Non-Null Count  Dtype  
---  ------                       --------------  -----  
 0   CountryID                    186 non-null    int64  
 1   Country Name                 186 non-null    object 
 2   WEBNAME                      186 non-null    object 
 3   Region                       186 non-null    object 
 4   World Rank                   180 non-null    float64
 5   Region Rank                  180 non-null    float64
 6   2019 Score                   180 non-null    float64
 7   Property Rights              185 non-null    float64
 8   Judical Effectiveness        185 non-null    float64
 9   Government Integrity         185 non-null    float64
 10  Tax Burden                   180 non-null    float64
 11  Gov't Spending               183 non-null    float64
 12  Fiscal Health                183 non-null    float64
 13  Business Freedom             185 non-null    float64
 14  Labor Freedom                184 non-null    float64
 15  Monetary Freedom             184 non-null    float64
 16  Trade Freedom                182 non-null    float64
 17  Investment Freedom           184 non-null    float64
 18  Financial Freedom            181 non-null    float64
 19  Tariff Rate (%)              182 non-null    float64
 20  Income Tax Rate (%)          183 non-null    float64
 21  Corporate Tax Rate (%)       183 non-null    float64
 22  Tax Burden % of GDP          179 non-null    float64
 23  Gov't Expenditure % of GDP   182 non-null    float64
 24  Country                      186 non-null    object 
 25  Population (Millions)        186 non-null    object 
 26  GDP (Billions, PPP)          185 non-null    object 
 27  GDP Growth Rate (%)          184 non-null    float64
 28  5 Year GDP Growth Rate (%)   183 non-null    float64
 29  GDP per Capita (PPP)         184 non-null    object 
 30  Unemployment (%)             181 non-null    object 
 31  Inflation (%)                182 non-null    float64
 32  FDI Inflow (Millions)        181 non-null    object 
 33  Public Debt (% of GDP)       182 non-null    float64
dtypes: float64(24), int64(1), object(9)
memory usage: 49.5+ KB


freedom_df.columns = freedom_df.columns.str.replace(' ', '_')


# Check missing values
def missing_values(df):
    miss_val = df.isnull().sum()
    miss_val_percent = df.isnull().sum() / len(df) * 100
    cols = freedom_df.columns.values
    sort_ = np.argsort(cols)
    indices = pd.Series(sort_[np.searchsorted(cols,cols,sorter=sort_)],
                       index=cols)
    miss_val_table = pd.concat([indices, miss_val, miss_val_percent],
                               axis=1)
    miss_val_df = miss_val_table.rename(
    columns = {0: "Column Index", 1: "Missing Values", 2: "% Missing"})
    return miss_val_df


missing_values(freedom_df)


na_count = freedom_df.isnull().sum(axis=1)
na_data = na_count[na_count > 0]
pd.concat([freedom_df.loc[na_data.index]['Country_Name'],na_data], 
          axis=1).rename(columns={0: "NA Count"})


freedom_df.drop( freedom_df[ freedom_df['Country_Name'] == 'Liechtenstein' ].index, inplace=True)
freedom_df.drop( freedom_df[ freedom_df['Country_Name'] == 'Somalia' ].index, inplace=True)
freedom_df.drop( freedom_df[ freedom_df['Country_Name'] == 'Syria' ].index, inplace=True)


cols_to_consider=['2019_Score', 'Property_Rights', 'Judical_Effectiveness',
       'Government_Integrity', 'Tax_Burden',"Gov't_Spending", 'Fiscal_Health', 'Business_Freedom'
        ,'Labor_Freedom', 'Monetary_Freedom', 'Trade_Freedom','Investment_Freedom_'
        ,'Financial_Freedom', 'Tariff_Rate_(%)',
       'Income_Tax_Rate_(%)', 'Corporate_Tax_Rate_(%)', 'Tax_Burden_%_of_GDP',
       "Gov't_Expenditure_%_of_GDP_", 'Population_(Millions)',
       'GDP_(Billions,_PPP)', 'GDP_Growth_Rate_(%)', 'GDP_per_Capita_(PPP)',
       'Unemployment_(%)', 'Inflation_(%)', 'FDI_Inflow_(Millions)',
       'Public_Debt_(%_of_GDP)']


subset_df=freedom_df[cols_to_consider]  #Selecting only the above columns


#Checking for missing values
subset_df.isna().sum()

2019_Score                     3
Property_Rights                0
Judical_Effectiveness          0
Government_Integrity           0
Tax_Burden                     3
Gov't_Spending                 0
Fiscal_Health                  0
Business_Freedom               0
Labor_Freedom                  0
Monetary_Freedom               0
Trade_Freedom                  2
Investment_Freedom_            1
Financial_Freedom              3
Tariff_Rate_(%)                2
Income_Tax_Rate_(%)            2
Corporate_Tax_Rate_(%)         2
Tax_Burden_%_of_GDP            4
Gov't_Expenditure_%_of_GDP_    1
Population_(Millions)          0
GDP_(Billions,_PPP)            0
GDP_Growth_Rate_(%)            0
GDP_per_Capita_(PPP)           0
Unemployment_(%)               5
Inflation_(%)                  1
FDI_Inflow_(Millions)          3
Public_Debt_(%_of_GDP)         1
dtype: int64


l_df = len(subset_df)
# Remove missing data
subset_df = subset_df.dropna()


print("%s Percent of Original Data Remaining" 
      %round(len(subset_df) / l_df * 100, 2))

94.54 Percent of Original Data Remaining


subset_df.shape

(173, 26)


subset_df.describe()


# Remove Special Characters from the columns and converting str to float
subset_df['GDP_(Billions,_PPP)']=subset_df['GDP_(Billions,_PPP)'].str.replace('$','').astype(str)
subset_df['GDP_(Billions,_PPP)']=subset_df['GDP_(Billions,_PPP)'].str.replace(',','').astype(float)

subset_df['GDP_per_Capita_(PPP)']=subset_df['GDP_per_Capita_(PPP)'].str.replace('$','').astype(str)
subset_df['GDP_per_Capita_(PPP)']=subset_df['GDP_per_Capita_(PPP)'].str.replace(',','').astype(float)


subset_df['FDI_Inflow_(Millions)']=subset_df['FDI_Inflow_(Millions)'].str.replace(',','').astype(float)


subset_df.head()


#Check for missing values
subset_df.isnull().sum()

2019_Score                     0
Property_Rights                0
Judical_Effectiveness          0
Government_Integrity           0
Tax_Burden                     0
Gov't_Spending                 0
Fiscal_Health                  0
Business_Freedom               0
Labor_Freedom                  0
Monetary_Freedom               0
Trade_Freedom                  0
Investment_Freedom_            0
Financial_Freedom              0
Tariff_Rate_(%)                0
Income_Tax_Rate_(%)            0
Corporate_Tax_Rate_(%)         0
Tax_Burden_%_of_GDP            0
Gov't_Expenditure_%_of_GDP_    0
Population_(Millions)          0
GDP_(Billions,_PPP)            0
GDP_Growth_Rate_(%)            0
GDP_per_Capita_(PPP)           0
Unemployment_(%)               0
Inflation_(%)                  0
FDI_Inflow_(Millions)          0
Public_Debt_(%_of_GDP)         0
dtype: int64


#Univariate analysis
for i in subset_df.columns:
    sns.distplot(subset_df[i])
    plt.show()

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)

C:\Users\anil.devireddy\Anaconda3\lib\site-packages\seaborn\distributions.py:2551: FutureWarning: `distplot` is a deprecated function and will be removed in a future version. Please adapt your code to use either `displot` (a figure-level function with similar flexibility) or `histplot` (an axes-level function for histograms).
  warnings.warn(msg, FutureWarning)


#Bivariate Analysis
sns.pairplot(subset_df , diag_kind = 'kde')
plt.show()


for i, name in enumerate(subset_df.columns):
    print(i, name)

0 2019_Score
1 Property_Rights
2 Judical_Effectiveness
3 Government_Integrity
4 Tax_Burden
5 Gov't_Spending
6 Fiscal_Health
7 Business_Freedom
8 Labor_Freedom
9 Monetary_Freedom
10 Trade_Freedom
11 Investment_Freedom_
12 Financial_Freedom
13 Tariff_Rate_(%)
14 Income_Tax_Rate_(%)
15 Corporate_Tax_Rate_(%)
16 Tax_Burden_%_of_GDP
17 Gov't_Expenditure_%_of_GDP_
18 Population_(Millions)
19 GDP_(Billions,_PPP)
20 GDP_Growth_Rate_(%)
21 GDP_per_Capita_(PPP)
22 Unemployment_(%)
23 Inflation_(%)
24 FDI_Inflow_(Millions)
25 Public_Debt_(%_of_GDP)


subset_df.isna().sum()

2019_Score                     0
Property_Rights                0
Judical_Effectiveness          0
Government_Integrity           0
Tax_Burden                     0
Gov't_Spending                 0
Fiscal_Health                  0
Business_Freedom               0
Labor_Freedom                  0
Monetary_Freedom               0
Trade_Freedom                  0
Investment_Freedom_            0
Financial_Freedom              0
Tariff_Rate_(%)                0
Income_Tax_Rate_(%)            0
Corporate_Tax_Rate_(%)         0
Tax_Burden_%_of_GDP            0
Gov't_Expenditure_%_of_GDP_    0
Population_(Millions)          0
GDP_(Billions,_PPP)            0
GDP_Growth_Rate_(%)            0
GDP_per_Capita_(PPP)           0
Unemployment_(%)               0
Inflation_(%)                  0
FDI_Inflow_(Millions)          0
Public_Debt_(%_of_GDP)         0
dtype: int64


scaler=StandardScaler()
subset_scaled=scaler.fit_transform(subset_df)


subset_scaled_df=pd.DataFrame(subset_scaled,columns=subset_df.columns)   #Creating a dataframe of the above results


#Identify the correlation between feature
sns.heatmap(subset_scaled_df.corr(),annot=True)
plt.show()


X = subset_df.drop('2019_Score', axis=1)

# Copy the 'mpg' column alone into the y dataframe. This is the dependent variable
y = subset_df[['2019_Score']]


from sklearn.linear_model import LinearRegression


lm = LinearRegression(normalize=True)
lm.fit(X,y)
lm.score(X,y)

0.9999927760468311


cols_to_consider1=['2019_Score', 'Property_Rights', 'Fiscal_Health', 'Business_Freedom'
        ,'Labor_Freedom', 'Monetary_Freedom', 'Trade_Freedom','Investment_Freedom_'
        ,'Financial_Freedom', 'Income_Tax_Rate_(%)', 'Tax_Burden_%_of_GDP',
       "Gov't_Expenditure_%_of_GDP_", 
       'GDP_(Billions,_PPP)', 'GDP_Growth_Rate_(%)', 
       'FDI_Inflow_(Millions)',
       'Public_Debt_(%_of_GDP)']


subset_df1=subset_df[cols_to_consider1]  #Selecting only the above columns


X = subset_df1.drop('2019_Score', axis=1)

# Copy the 'mpg' column alone into the y dataframe. This is the dependent variable
y = subset_df1[['2019_Score']]


lm = LinearRegression(normalize=True)
lm.fit(X,y)
lm.score(X,y)

0.9815666568438304


scaler=StandardScaler()
subset_scaled=scaler.fit_transform(subset_df1)


subset_scaled_df1=pd.DataFrame(subset_scaled,columns=subset_df1.columns)   #Creating a dataframe of the above results


sns.heatmap(subset_scaled_df1.corr(),annot=True)
plt.show()


cols_to_consider2=['2019_Score', 'Property_Rights', 'Fiscal_Health', 'Business_Freedom'
        ,'Labor_Freedom', 'Monetary_Freedom', 'Trade_Freedom','Investment_Freedom_'
        ,'Financial_Freedom']


subset_df2=subset_df[cols_to_consider2]  #Selecting only the above columns


X = subset_df2.drop('2019_Score', axis=1)

# Copy the 'mpg' column alone into the y dataframe. This is the dependent variable
y = subset_df2[['2019_Score']]


lm = LinearRegression(normalize=True)
lm.fit(X,y)
lm.score(X,y)

0.9491920166381992


scaler=StandardScaler()
subset_scaled=scaler.fit_transform(subset_df2)


subset_scaled_df2=pd.DataFrame(subset_scaled,columns=subset_df2.columns)   #Creating a dataframe of the above results


sns.heatmap(subset_scaled_df2.corr(),annot=True)
plt.show()


 subset_scaled_df2.drop('2019_Score', axis=1, inplace=True)


clusters=range(1,10)
meanDistortions=[]

for k in clusters:
    model=KMeans(n_clusters=k)
    model.fit(subset_scaled_df2)
    prediction=model.predict(subset_scaled_df2)
    distortion=sum(np.min(cdist(subset_scaled_df2, model.cluster_centers_, 'euclidean'), axis=1)) / subset_scaled_df2.shape[0]
                           
    meanDistortions.append(distortion)

    print(k,distortion)
plt.plot(clusters, meanDistortions, 'bx-')
plt.xlabel('k')
plt.ylabel('Average distortion')
plt.title('Selecting k with the Elbow Method')
plt.show()

1 2.6376796271077896
2 2.0779308877586593
3 1.958997063987088
4 1.832788077332468
5 1.7590807688748293
6 1.700942489593267
7 1.6276267114554284
8 1.5848344056316839
9 1.542974338949283


kmeans = KMeans(n_clusters=6, n_init = 15, random_state=2345)
kmeans.fit(subset_scaled_df2)

KMeans(n_clusters=6, n_init=15, random_state=2345)


centroids = kmeans.cluster_centers_
centroids

array([[ 2.97587899e-02,  5.93135223e-01, -4.86919899e-02,
        -4.46378325e-01,  3.14774229e-01,  5.89300209e-01,
         4.84169635e-01,  4.62467708e-01],
       [-6.52063216e-01, -1.35409621e+00, -5.30765389e-01,
        -5.26249118e-01, -2.71175366e-01, -7.38695207e-01,
        -3.83134765e-01, -4.40027466e-01],
       [ 1.48140595e+00,  6.24818324e-01,  1.19851203e+00,
         5.52776632e-01,  7.00275105e-01,  1.05241035e+00,
         1.06351812e+00,  1.19960776e+00],
       [-2.39714414e+00, -1.60172254e+00, -2.09277863e+00,
        -2.33298154e+00, -7.98836545e+00, -1.39167636e+00,
        -2.74448178e+00, -2.10176461e+00],
       [ 2.50843990e-01, -2.17050885e-01,  4.48533861e-01,
         9.95799575e-01,  1.33014931e-01,  6.70601458e-02,
         9.98063710e-02, -5.47106573e-04],
       [-9.55726629e-01,  3.99030360e-01, -8.84976219e-01,
        -2.54553523e-01, -6.24538264e-01, -9.08364433e-01,
        -1.17175774e+00, -1.16284220e+00]])


centroid_df = pd.DataFrame(centroids, columns = subset_scaled_df2.columns )


centroid_df


dataset=subset_scaled_df2[:]


dataset['KmeansLabel']=kmeans.labels_

<ipython-input-197-2f5e4caa51de>:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  dataset['KmeansLabel']=kmeans.labels_


dataset.head(10)


plt.scatter(dataset['Property_Rights'], dataset['Financial_Freedom'], c=kmeans.labels_,)  
plt.show()


plt.scatter(dataset['Trade_Freedom'], dataset['Investment_Freedom_'], c=kmeans.labels_,)  
plt.show()


plt.scatter(dataset['Trade_Freedom'], dataset['Business_Freedom'], c=kmeans.labels_,)  
plt.show()


#Hierarchial Clustering


linkage_methods=['single','complete','average','ward','median']
results_cophenetic_coef=[]
for i in linkage_methods :
    plt.figure(figsize=(15, 13))
    plt.xlabel('sample index')
    plt.ylabel('Distance')
    Z = linkage(subset_scaled_df2, i)
    cc,cophn_dist=cophenet(Z,pdist(subset_scaled_df2))
    dendrogram(Z,leaf_rotation=90.0,p=5,leaf_font_size=10,truncate_mode='level')
    plt.tight_layout()
    plt.title("Linkage Type: "+ i +" having cophenetic coefficient : "+str(round(cc,3)) )
    plt.show()
    results_cophenetic_coef.append((i,cc))
    print (i,cc)

single 0.640253242243292

complete 0.6209212797798624

average 0.7345934558949965

ward 0.4760173531768205

median 0.6100529126421079


results_cophenetic_coef_df=pd.DataFrame(results_cophenetic_coef,columns=['LinkageMethod','CopheneticCoefficient'])
results_cophenetic_coef_df


#use truncate_mode='lastp' to select last p formed clusters
plt.figure(figsize=(10,8))
Z = linkage(subset_scaled_df2, 'complete', metric='euclidean')

dendrogram(
    Z,
    truncate_mode='lastp',  # show only the last p merged clusters
    p=25 # show only the last p merged clusters
)
plt.show()


max_d=5
from scipy.cluster.hierarchy import fcluster
clusters = fcluster(Z, max_d, criterion='distance')


set(clusters)

{1, 2, 3, 4, 5, 6, 7, 8}


dataset2=subset_scaled_df2[:]


dataset2['HierarchicalClusteringLabel']=clusters

<ipython-input-216-f9881820bb96>:1: SettingWithCopyWarning: 
A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy
  dataset2['HierarchicalClusteringLabel']=clusters


dataset2.head(3)


from sklearn.metrics import silhouette_score
silhouette_score(dataset.drop('KmeansLabel',axis=1),dataset['KmeansLabel'])

0.19043170387074412


from sklearn.metrics import silhouette_score
silhouette_score(dataset2.drop('HierarchicalClusteringLabel',axis=1),dataset2['HierarchicalClusteringLabel'])

0.16311743235141693

	CountryID	Country Name	WEBNAME	Region	World Rank	Region Rank	2019 Score	Property Rights	Judical Effectiveness	Government Integrity	...	Country	Population (Millions)	GDP (Billions, PPP)	GDP Growth Rate (%)	5 Year GDP Growth Rate (%)	GDP per Capita (PPP)	Unemployment (%)	Inflation (%)	FDI Inflow (Millions)	Public Debt (% of GDP)
0	1	Afghanistan	Afghanistan	Asia-Pacific	152.0	39.0	51.5	19.6	29.6	25.2	...	Afghanistan	35.5	$69.6	2.5	2.9	$1,958	8.8	5.0	53.9	7.3
1	2	Albania	Albania	Europe	52.0	27.0	66.5	54.8	30.6	40.4	...	Albania	2.9	$36.0	3.9	2.5	$12,507	13.9	2.0	1,119.1	71.2
2	3	Algeria	Algeria	Middle East and North Africa	171.0	14.0	46.2	31.6	36.2	28.9	...	Algeria	41.5	$632.9	2.0	3.1	$15,237	10.0	5.6	1,203.0	25.8
3	4	Angola	Angola	Sub-Saharan Africa	156.0	33.0	50.6	35.9	26.6	20.5	...	Angola	28.2	$190.3	0.7	2.9	$6,753	8.2	31.7	-2,254.5	65.3
4	5	Argentina	Argentina	Americas	148.0	26.0	52.2	47.8	44.5	33.5	...	Argentina	44.1	$920.2	2.9	0.7	$20,876	8.7	25.7	11,857.0	52.6

	2019_Score	Property_Rights	Judical_Effectiveness	Government_Integrity	Tax_Burden	Gov't_Spending	Fiscal_Health	Business_Freedom	Labor_Freedom	Monetary_Freedom	...	Investment_Freedom_	Financial_Freedom	Tariff_Rate_(%)	Income_Tax_Rate_(%)	Corporate_Tax_Rate_(%)	Tax_Burden_%_of_GDP	Gov't_Expenditure_%_of_GDP_	GDP_Growth_Rate_(%)	Inflation_(%)	Public_Debt_(%_of_GDP)
count	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	...	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000	173.000000
mean	61.349133	53.561850	46.045087	42.284393	77.704046	66.019075	66.811561	64.722543	60.051445	75.682659	...	58.583815	49.653179	5.623699	28.604046	23.916763	22.045665	32.023121	3.290751	10.869942	56.667052
std	10.290731	19.229243	17.651070	19.925058	11.786874	20.957233	30.813333	14.770800	13.778284	9.501612	...	21.408001	18.921378	4.408220	13.235103	8.878332	10.151558	10.387187	2.748959	82.555305	33.854417
min	25.900000	7.600000	12.300000	7.900000	42.000000	0.900000	0.000000	17.700000	20.000000	0.000000	...	0.000000	10.000000	0.000000	0.000000	0.000000	1.600000	10.600000	-14.000000	-0.900000	0.000000
25%	54.600000	37.400000	32.300000	28.100000	71.000000	54.100000	42.500000	55.800000	51.100000	72.400000	...	45.000000	40.000000	2.000000	20.000000	20.000000	14.000000	23.600000	2.000000	1.300000	35.200000
50%	60.800000	52.400000	44.000000	36.400000	78.600000	70.000000	80.300000	65.800000	60.100000	77.800000	...	60.000000	50.000000	4.200000	30.000000	25.000000	20.700000	31.600000	3.200000	2.800000	50.400000
75%	68.100000	66.000000	56.000000	50.500000	85.400000	83.300000	90.100000	75.200000	69.200000	81.800000	...	75.000000	60.000000	8.700000	35.000000	30.000000	29.500000	39.100000	4.800000	5.500000	70.200000
max	90.200000	97.400000	92.400000	96.700000	99.800000	96.600000	100.000000	96.400000	91.000000	88.000000	...	95.000000	90.000000	18.600000	60.000000	50.000000	47.000000	57.500000	10.900000	1087.500000	236.400000

	2019_Score	Property_Rights	Judical_Effectiveness	Government_Integrity	Tax_Burden	Gov't_Spending	Fiscal_Health	Business_Freedom	Labor_Freedom	Monetary_Freedom	...	Tax_Burden_%_of_GDP	Gov't_Expenditure_%_of_GDP_	Population_(Millions)	GDP_(Billions,_PPP)	GDP_Growth_Rate_(%)	GDP_per_Capita_(PPP)	Unemployment_(%)	Inflation_(%)	FDI_Inflow_(Millions)	Public_Debt_(%_of_GDP)
0	51.5	19.6	29.6	25.2	91.7	80.3	99.3	49.2	60.4	76.7	...	5.0	25.6	35.5	69.6	2.5	1958.0	8.8	5.0	53.9	7.3
1	66.5	54.8	30.6	40.4	86.3	73.9	80.6	69.3	52.7	81.5	...	24.9	29.5	2.9	36.0	3.9	12507.0	13.9	2.0	1119.1	71.2
2	46.2	31.6	36.2	28.9	76.4	48.7	18.7	61.6	49.9	74.9	...	24.5	41.4	41.5	632.9	2.0	15237.0	10.0	5.6	1203.0	25.8
3	50.6	35.9	26.6	20.5	83.9	80.7	58.2	55.7	58.8	55.4	...	20.6	25.3	28.2	190.3	0.7	6753.0	8.2	31.7	-2254.5	65.3
4	52.2	47.8	44.5	33.5	69.3	49.5	33.0	56.4	46.9	60.2	...	30.8	41.0	44.1	920.2	2.9	20876.0	8.7	25.7	11857.0	52.6

	Property_Rights	Fiscal_Health	Business_Freedom	Labor_Freedom	Monetary_Freedom	Trade_Freedom	Investment_Freedom_	Financial_Freedom
0	0.029759	0.593135	-0.048692	-0.446378	0.314774	0.589300	0.484170	0.462468
1	-0.652063	-1.354096	-0.530765	-0.526249	-0.271175	-0.738695	-0.383135	-0.440027
2	1.481406	0.624818	1.198512	0.552777	0.700275	1.052410	1.063518	1.199608
3	-2.397144	-1.601723	-2.092779	-2.332982	-7.988365	-1.391676	-2.744482	-2.101765
4	0.250844	-0.217051	0.448534	0.995800	0.133015	0.067060	0.099806	-0.000547
5	-0.955727	0.399030	-0.884976	-0.254554	-0.624538	-0.908364	-1.171758	-1.162842

	Property_Rights	Fiscal_Health	Business_Freedom	Labor_Freedom	Monetary_Freedom	Trade_Freedom	Investment_Freedom_	Financial_Freedom	KmeansLabel
0	-1.771283	1.057424	-1.053944	0.025371	0.107381	-0.835841	-2.276011	-2.101765	5
1	0.064576	0.448782	0.310799	-0.535102	0.614025	1.183693	0.534815	1.078456	0
2	-1.145422	-1.565920	-0.212013	-0.738910	-0.082610	-0.706146	-1.339069	-1.041691	1
3	-0.921155	-0.280286	-0.612610	-0.091091	-2.140851	-1.280509	-1.339069	-0.511654	5
4	-0.300510	-1.100488	-0.565081	-0.957276	-1.634207	-0.465284	-0.167891	0.548420	1
5	0.189748	-0.449534	0.921878	0.826046	0.223487	0.535219	0.769051	1.078456	4
6	1.331944	0.631049	1.600854	1.750462	1.152334	1.165165	1.003286	2.138530	2
7	1.597935	0.608265	0.691026	0.629517	0.614025	1.016943	1.471757	1.078456	2
8	0.288843	0.735201	0.324378	0.280131	-1.338665	-0.039144	0.066344	0.548420	4
9	-0.592578	-0.036179	0.256480	0.542170	0.255152	-2.521875	-0.402127	0.548420	1

	Column Index	Missing Values	% Missing
CountryID	0	0	0.000000
Country_Name	1	0	0.000000
WEBNAME	2	0	0.000000
Region	3	0	0.000000
World_Rank	4	6	3.225806
Region_Rank	5	6	3.225806
2019_Score	6	6	3.225806
Property_Rights	7	1	0.537634
Judical_Effectiveness	8	1	0.537634
Government_Integrity	9	1	0.537634
Tax_Burden	10	6	3.225806
Gov't_Spending	11	3	1.612903
Fiscal_Health	12	3	1.612903
Business_Freedom	13	1	0.537634
Labor_Freedom	14	2	1.075269
Monetary_Freedom	15	2	1.075269
Trade_Freedom	16	4	2.150538
Investment_Freedom_	17	2	1.075269
Financial_Freedom	18	5	2.688172
Tariff_Rate_(%)	19	4	2.150538
Income_Tax_Rate_(%)	20	3	1.612903
Corporate_Tax_Rate_(%)	21	3	1.612903
Tax_Burden_%_of_GDP	22	7	3.763441
Gov't_Expenditure_%_of_GDP_	23	4	2.150538
Country	24	0	0.000000
Population_(Millions)	25	0	0.000000
GDP_(Billions,_PPP)	26	1	0.537634
GDP_Growth_Rate_(%)	27	2	1.075269
5_Year_GDP_Growth_Rate_(%)	28	3	1.612903
GDP_per_Capita_(PPP)	29	2	1.075269
Unemployment_(%)	30	5	2.688172
Inflation_(%)	31	4	2.150538
FDI_Inflow_(Millions)	32	5	2.688172
Public_Debt_(%_of_GDP)	33	4	2.150538

	Country_Name	NA Count
42	Cuba	1
47	Dominica	1
78	Iraq	9
87	Kiribati	1
88	Korea, North	6
90	Kosovo	1
98	Libya	9
99	Liechtenstein	21
113	Micronesia	2
147	Seychelles	1
153	Somalia	19
161	Syria	16
183	Yemen	9

	LinkageMethod	CopheneticCoefficient
0	single	0.640253
1	complete	0.620921
2	average	0.734593
3	ward	0.476017
4	median	0.610053